--- title: "Iluvatar CoreX Unveils Tiangai 300 Benchmarks: Chinese Accelerator Claimed on Par with Nvidia H200" description: "China's Iluvatar CoreX showcased Tiangai 300 benchmarks to journalists, claiming competition with the Nvidia H200 and local superiority in AI tasks. Comparison methodology was not fully disclosed." date: 2026-10-31T17:28:38.031Z lang: en url: https://xab.info/en/posts/iluvatar-corex-tiangai-300-nvidia-h200 tags: [iluvatar-corex, nvidia-h200, ai-accelerator, china-chips, tiangai-300, gpu] author: "Alexander Buran" publisher: "XAB.info" --- # Iluvatar CoreX Unveils Tiangai 300 Benchmarks: Chinese Accelerator Claimed on Par with Nvidia H200 ![Modern Iluvatar CoreX Tiangai 300 graphics accelerator set against a futuristic blue glowing background](https://xab.info/media/2026/10/31/iluvatar-tiangai-300-nvidia-h200/iluvatar-tiangai-300-nvidia-h200-1.webp) ## Key Points - Iluvatar CoreX stated that the Tiangai 300 competes with the Nvidia H200 and outperforms it in selected AI tasks. - In the image generation test, the Tiangai 300 system completed 100% of requests compared to 87.5% for the H200, though with a longer completion time. - INESA testing showed a 5% higher throughput per accelerator for the Tiangai 300 in a prefill/decode configuration. - The company aims to surpass the Nvidia Rubin platform within two years; Tiangai is already deployed in systems scaling to tens of thousands of cards. - Methodology and criteria for some tests are not fully disclosed, and results apply to specific workloads. Chinese company Iluvatar CoreX has announced that its general-purpose graphics accelerator, the Tiangai 300, is capable of competing with the Nvidia H200—Nvidia's most advanced AI accelerator permitted for export to China under current US regulations. During a demonstration in Shanghai, the manufacturer presented benchmark results to journalists and showcased the hardware in operation. However, the company notes that detailed comparison conditions and calculation methodologies for some metrics have not been fully disclosed, limiting the possibility of independent verification of the claimed figures. ### Image Generation: 100% vs. 87.5% In one text-to-image generation test, the Tiangai 300 system successfully completed 100% of requests with a throughput of 0.101 images per second, while the H200 system handled 87.5% of requests at 0.086 images per second. At the same time, the Chinese accelerator's task completion time was 0.64 seconds longer. The company does not specify the criteria for successful request completion or the calculation methodology, so a direct comparison of these figures requires caution. ### Third-Party Testing and Prefill/Decode Configuration According to Iluvatar CoreX, testing conducted in July by Shanghai-based INESA showed that the Tiangai 300 system delivers 5% higher throughput per accelerator than the H200 system. The tests were run in a configuration with separated processing of input requests (prefill) and response generation (decode). The manufacturer claims the Tiangai 300 is the first Chinese general-purpose GPU to reach the H200 level in such tests. ### LLM Workloads and Model Training In another test, the company claims the Tiangai 300 outperformed the H200 by 15% in throughput when running the GLM-5.3 model on a Mixture-of-Experts (MoE) architecture with a 64K token context. For AI agent tasks with contexts ranging from 8K to 128K tokens, time-to-first-token was 10–30% lower. In a demonstration of training the open Alibaba Qwen3-32B model, an eight-card Tiangai 300 system using FP8 precision delivered 1.33 times higher effective throughput per card than an Nvidia H100 system. ### Roadmap and Deployment Scale Iluvatar CoreX is steadily closing the gap with Nvidia: in January, the company published a long-term architecture roadmap stating its intention to surpass the Nvidia Rubin platform within two years. According to the manufacturer, Tiangai series accelerators are already deployed in systems numbering tens of thousands of cards, accumulating valuable deployment and maintenance experience. The company is also expanding beyond cloud data centers to develop edge AI solutions: the Tongyang TY1200 computing system outperformed an Nvidia Thor T5000-based solution by 20% in certain large language model tests. ### Limitations of Claimed Results It is important to note that all cited figures pertain to selected workloads and do not prove overall superiority of the Tiangai 300 over its competitor. Success criteria, configurations, and methodologies for some tests have not been fully disclosed, and some tests were conducted by the company or its partner itself. Therefore, Iluvatar CoreX's statement should be viewed as a significant yet not fully verifiable step by the Chinese industry toward parity with Nvidia's flagship AI accelerators. ## Fact-Check & Sources - [Китайская Iluvatar заявила, что сделала ИИ-ускоритель не хуже Nvidia H200 — и даже показала тесты window-new](https://3dnews.ru/1150668/kitayskaya-iluvatar-zayavila-chto-sdelala-iiuskoritel-ne-huge-nvidia-h200-i-dage-pokazala-testi) - Primary source: all test figures, Shanghai demonstration, INESA testing, roadmap, and caveats regarding incomplete disclosure of methodology. - [Китайская Iluvatar заявила, что сделала ИИ-ускоритель не хуже Nvidia H200](https://runet.news/news/86117) - Rebroadcast of the same material; facts match the primary source. ## FAQ ### What is the Tiangai 300? It is a general-purpose graphics accelerator (GPU) by Chinese company Iluvatar CoreX, positioned by the manufacturer as a competitor to the Nvidia H200 for artificial intelligence workloads. ### What results were shown during the Shanghai demonstration? In the image generation test, the Tiangai 300 system completed 100% of requests with a throughput of 0.101 images/sec, while the H200 system achieved 87.5% at 0.086 images/sec, though Tiangai's completion time was 0.64 seconds longer. ### Are the results verified by an independent third party? Partially: testing was conducted by Shanghai-based INESA, showing that the Tiangai 300 provides 5% higher throughput per accelerator in a prefill/decode configuration. Full methodology and criteria for some tests have not been disclosed. ### What are Iluvatar CoreX's future plans? The company published a roadmap aiming to surpass the Nvidia Rubin platform within two years and states that Tiangai accelerators are already used in systems scaling to tens of thousands of cards.